Managing satellite precipitation data (PERSIANN) through web geoservices: A case study in North Vietnam

Abstract

Rainfall is one of the most important factors affecting various types of hazards such as: landslides, floods, sea level rise and so on. With the availability of satellite rainfall estimates at fine time and space resolution, it has also become possible to mitigate such problems over the world. But satellite rainfall needs to be monitored before use, because the satellite data does not reflect the strong influences on precipitation of topography in some cases. Relief of study area is very complex including mountain and plain areas. In this paper we present a Decision System and an intelligent geoportal for North Vietnam based on Web Service allowing users to investigate satellite rainfall by means of a direct comparison and of the Revised Universal Soil Loss Equation (RUSLE) model. The comparison method uses data from Precipitation Estimation from Remote Sensing Information using Artificial Neural Network (PERSIANN) and rain gauges (RG) to investigate the interpolation of RG data. Furthermore, we also estimate a correlation and examined a percentage of simultaneous rain or no-rain between them. We realize that correlation between PERSIANN and gauge data meets expectation value when we investigate monthly data. The RUSLE model for computing the soil loss, which requires a huge amount of information and data, was handled for both PERSIANN product and rain gauges data to estimate the difference due to the usage of the two data sources.